Numerical simulation method for sudden opening of building based on porous medium and pressure change under sudden opening

Abstract

The application discloses a numerical simulation method for building sudden opening based on a porous medium, and comprises the following steps: S1, a building model is established by using modeling software, openings are arranged on the windward surface of the building model, and the building model is divided into grids; S2, a porous medium area is defined at the openings of the building model, and the porous medium area covers the openings of the building model; and S3, a resistance source term is added to the porous medium area, and the resistance coefficient is greater than or equal to a set threshold value under initial conditions; during the numerical simulation process, the resistance coefficient is suddenly equal to zero or is reduced to zero according to a set curve, so that the simulation effect of the sudden opening is achieved. The application further discloses a numerical simulation method for internal and external pressure changes under building sudden opening. The numerical simulation method for building sudden opening based on the porous medium and the numerical simulation method for internal and external pressure changes under building sudden opening can realize dynamic simulation of building sudden opening and numerical simulation of internal and external pressure changes under building sudden opening.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of civil and hydraulic engineering, and particularly relates to a numerical simulation method for internal and external pressure changes of a building with a sudden opening and the sudden opening. BACKGROUND

[0002] In recent years, wind disasters have occurred frequently in coastal areas of China. Post-disaster data shows that the roofs and other enclosure structures of low-rise buildings are severely damaged, accounting for more than half of the total losses. Such damage is often caused by flying objects carried by strong winds damaging the doors and windows of the building, causing the internal air pressure of the building to rise sharply, and the combined effect of the internal and external pressures causing the damage of the roof and other enclosure structures of the building. In the study of the internal and external pressure responses of wind-induced opening structures, a large number of scholars have used wind tunnel test methods to study the different opening parameters (opening number, opening position, opening shape, etc.) of buildings. At the same time, some scholars have also studied the transient internal pressure response of a closed model with a sudden opening. However, the existing research is mainly based on wind tunnel tests, and lacks numerical simulation research under the corresponding transient opening working condition, resulting in a lack of more in-depth discussion of the sudden opening of buildings. The key problem causing this phenomenon is that the existing numerical simulation method is difficult to simulate the dynamic opening process of the building.

[0003] A porous medium, defined as a substance with numerous small pores, has been widely used to simulate the seepage motion of gas or liquid between solids with different porosities. In numerical simulation, the flow of fluid through a porous medium solid can be effectively simulated by artificially defining a specific grid area and adding a resistance source term. This method is not only suitable for simulating seepage motion in the field of geotechnical engineering, but also is extended to simulate natural objects such as tree crowns and tree trunks to accurately reproduce the flow field distribution of wind penetrating through trees. Importantly, the porosity of the porous medium plays a key role in achieving this simulation process. When the porosity is set within a reasonable range, the geometric model can accurately simulate real porous substances such as trees and soil. In particular, when the porosity is set to 0%, the porous medium model will tend to be a geometric entity. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a numerical simulation method for internal and external pressure changes of a building with a sudden opening based on a porous medium, which can realize dynamic simulation of the sudden opening of a building and numerical simulation of the internal and external pressure changes under the sudden opening of the building.

[0005] To achieve the above purpose, the present application provides the following technical solutions:

[0006] The present application first proposes a numerical simulation method for the sudden opening of a building based on a porous medium, including the following steps:

[0007] S1: establishing a building model by using modeling software, setting an opening on the windward surface of the building model and dividing a grid;

[0008] S2: defining a porous medium region at the opening of the building model and covering the opening of the building model with the porous medium region;

[0009] S3: adding a resistance source term to the porous medium region, so that the resistance coefficient is greater than or equal to a set threshold value under initial conditions; during numerical simulation, the resistance coefficient is suddenly equal to zero or decreases to zero according to a set curve, so as to achieve the simulation effect of sudden opening.

[0010] Further, in the step S2, a porous medium block model with the same size as the opening is placed at the porous medium region.

[0011] Further, the step S3 comprises the following steps:

[0012] 131) adding a resistance source term: the porous medium block model is obtained based on the Darcy-Forchheimer theory, and the momentum equation of the porous medium block model after adding the resistance source term is:

[0013]

[0014] Wherein: S m represents the momentum source term in the m (m=x, y, z) direction; μ represents the fluid dynamic viscosity; ρ represents the fluid density; σ represents the stress tensor; U m represents the fluid velocity in the m (m=x, y, z) direction; |U| represents the velocity matrix in three directions; D and F are the viscous resistance coefficient D and the inertial resistance coefficient F in the Darcy-Forchheimer model, respectively;

[0015] 132) sudden opening: the viscous resistance coefficient D and the inertial resistance coefficient F are greater than or equal to a set threshold value under initial conditions; during numerical simulation, the viscous resistance coefficient D and the inertial resistance coefficient F are suddenly equal to zero or decrease to zero according to a set curve, so as to achieve the simulation effect of sudden opening.

[0016] The application further provides a numerical simulation method for internal and external pressure changes of a building with sudden opening, comprising the following steps:

[0017] Step 1: constructing a building model with sudden opening

[0018] 11) establishing a square building model by using modeling software, opening the building model on the windward surface and dividing a grid;

[0019] 12) defining a porous medium region at the opening of the building model and covering the opening of the building model with the porous medium region;

[0020] 13) adding a resistance source term in the porous medium region;

[0021] Step two: design the computational domain

[0022] A computational domain with size of 30Hx10Hx6H is constructed, the building model is placed in the computational domain, and the origin of the computational domain is located at the center of the bottom of the building model, the distance between the center of the bottom of the building model and the inlet and outlet of the computational domain is 10H and 20H respectively, and H is the height of the building model;

[0023] Step three: divide the grid

[0024] The building model and the computational domain are respectively divided into grids, and a boundary layer region is set at the wall surface of the building model to ensure that the boundary flow near the wall surface is accurately solved;

[0025] Step four: set the boundary conditions

[0026] Inlet boundary condition: the inlet adopts an atmospheric inflow boundary condition, and a uniform discrete inflow turbulence generation method is used;

[0027] Outlet boundary condition: the outlet boundary condition is set as a pressure outlet;

[0028] The top and both sides of the computational domain are set as slip boundary conditions, and the bottom of the computational domain and the building wall boundary condition are set as no-slip boundary conditions;

[0029] Step five: numerical simulation

[0030] Based on the large eddy turbulence model, under the initial condition, the resistance coefficient of the building model porous medium region is greater than or equal to the set threshold value under the initial condition to achieve the purpose of closing the opening; when the wind speed and wind pressure of the whole flow field are stable, the resistance coefficient of the building model porous medium region is set to 0 to achieve the purpose of opening the opening; the sudden opening simulation of the building is realized;

[0031] Step six: numerical analysis

[0032] The pressure and wind speed data inside and outside the building collected under the sudden opening of the building model are subjected to numerical analysis.

[0033] Further, in step 12), a porous medium block model with the same size as the opening is placed at the porous medium region;

[0034] The step 13) further comprises the following steps:

[0035] S31: Adding resistance source term: The porous medium block model is constructed based on the Darcy-Forchheimer theory, and the momentum equation of the porous medium block model after adding the resistance source term is:

[0036]

[0037] Wherein: S m represents the momentum source term in the m (m = x, y, z) direction; μ is the fluid dynamic viscosity; ρ represents the fluid density; σ represents the stress tensor; U m represents the fluid velocity in the m (m = x, y, z) direction; |U| represents the velocity matrix in three directions; D and F are the viscous resistance coefficient and inertial resistance coefficient in the Darcy-Forchheimer model, respectively;

[0038] S32: Sudden opening: make the viscous resistance coefficient D and the inertial resistance coefficient F greater than or equal to the set threshold value under the initial condition; in the numerical simulation process, make the viscous resistance coefficient D and the inertial resistance coefficient F both suddenly equal to zero or decrease to zero according to the set curve, so as to achieve the simulation effect of sudden opening.

[0039] Further, in step four, the mathematical form expression of the fluctuating wind speed synthesized by the uniform discrete flow turbulence generation method is:

[0040]

[0041] Wherein: u i , i = 1, 2, 3 represent the fluctuating wind speed in x, y, z three directions respectively; j = 1, 2, 3 represent x, y, z three directions respectively; M represents the number of division interval segments of the spectrum; N represents the number of frequencies in each interval segment; And are coefficients; represents a random vector subject to distribution on the spatial unit sphere, which ensures that the velocity field meets the divergence-free condition; represents the dimensionless coordinate vector, and is the correlation coefficient between two points in space; f n,m is the frequency, subject to uniform distribution with mean f m and standard deviation Δf = 0.1; t is time; and:

[0042]

[0043] Wherein: sign is the sign function; represents a random number subject to 0-1 normal distribution in three directions; S ui (f m ) is the spectrum f m corresponding to the power spectral density value in the i direction.

[0044] Furthermore, in the consistent discrete incoming flow turbulence generation method, an iterative reshaping method is introduced to continuously correct the shape of the wind speed profile at the inlet. This ensures that the turbulent wind field can fit the target wind profile within the target region without affecting spatial correlation, while guaranteeing the conservation of flow rate through the inlet interface. The principle of the iterative reshaping method is as follows:

[0045]

[0046] In the formula: U mean,t (z) and I i,t (z) represents the target mean velocity profile and the target turbulence intensity profile; and It is the average wind speed profile and turbulence intensity profile obtained from a simulated time history at the center of the building area during the nth correction period; and These are the mean wind speed profile and turbulence intensity profile of the turbulent field generated at the inlet interface during the nth correction; γ n Δz is a correction factor used to ensure that the flow rate through the inlet interface remains constant after the (n+1)th correction of the average wind speed profile at the inlet interface; Δz is the grid scale corresponding to the z-direction.

[0047] Furthermore, in step five, the method for simulating and solving the large eddy turbulence model is as follows:

[0048] Filter all vortex scales Φ(x,t) in the flow field:

[0049]

[0050] in: For large-scale vortices obtained through filtering; Φ′(x,t) is the subgrid scale; G(x,x′) is the spatial filtering function; and:

[0051]

[0052] Where: x is the spatial coordinate before filtering; x′ is the spatial coordinate after filtering; Δ is the filtering scale;

[0053] The filtered incompressible fluid control equations are as follows:

[0054]

[0055] in: (i, j = 1, 2, 3) are the three filtered velocity components in the Cartesian coordinate system; ρ is the filtered pressure; ν is the air density; ν is the kinematic viscosity; τ is the pressure after filtering. ij Represents subgrid-scale stress;

[0056] Small-scale vortices are modeled using a subgrid-based eddy viscosity model, where the expression for the standard SGS model is:

[0057]

[0058] Where: μ t It is the SGS eddy viscosity coefficient, δ ij It is the Kronecker function. It is the filter thickness strain rate tensor;

[0059] Eddy viscosity coefficient μ t Calculated using the Smagorinsky-Lilly model:

[0060]

[0061] Where: k is the von Karman constant; C s Δ is the Smagorinsky constant; Δ is the filtering scale.

[0062] The beneficial effects of this invention are as follows:

[0063] This invention presents a numerical simulation method for sudden openings in buildings based on porous media. By defining a porous media region covering the opening in the building model and adding a resistance source term within this region, the method allows for dynamic numerical simulation of sudden openings in buildings. This is achieved by adjusting the resistance coefficient of the porous media region during the numerical simulation of pressure changes under transient opening conditions. Furthermore, by adjusting the porosity, the method transforms a closed model into an open model. In the numerical simulation software, the sudden opening process of a closed model is innovatively realized through dynamic adjustment of the porous media porosity. This invention provides new possibilities for simulating complex fluid dynamics and interactions, and has broad application prospects. Attached Figure Description

[0064] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the following figures are provided for illustration:

[0065] Figure 1 This is a flowchart of the numerical simulation method for sudden openings in buildings based on porous media according to the present invention;

[0066] Figure 2 This is a two-dimensional mesh diagram of a building model;

[0067] Figure 3 The instantaneous velocity and pressure fields before and after the opening are shown.

[0068] Figure 4 Time history of wind pressure and wind speed at the openings inside the building;

[0069] Figure 5 A schematic diagram of the building model;

[0070] Figure 6 This is a schematic diagram of the computational domain and model location;

[0071] Figure 7 A schematic diagram of a computational domain encrypted mesh;

[0072] Figure 8 A schematic diagram of a dense mesh and a local mesh for a model with closed or open openings;

[0073] Figure 9 This is a wind field characteristic map;

[0074] Figure 10 A diagram showing the layout of measurement points for the model;

[0075] Figure 11 Diagram showing the layout of external and internal pressure measuring points for the opening;

[0076] Figure 12 Comparison of theoretical and simulated values ​​for average internal pressure;

[0077] Figure 13 A cloud map showing the average wind pressure coefficient at different opening ratios under a 0° angle;

[0078] Figure 14 Cloud maps of fluctuating wind pressure coefficients at different opening ratios at a 0° angle;

[0079] Figure 15 The wind pressure distribution on the central axis of the model under different aperture ratios at a 0° angle;

[0080] Figure 16 The external average wind pressure is given by the open and closed models at a 45° angle.

[0081] Figure 17 The external pulsating wind pressure is given for the open and closed models at a 45° angle. Detailed Implementation

[0082] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.

[0083] Example 1: Numerical Simulation Method for Sudden Openings in Buildings Based on Porous Media

[0084] like Figure 1 As shown in the figure, the numerical simulation method for sudden openings in buildings based on porous media in this embodiment includes the following steps:

[0085] S1: Use modeling software to create a building model, set openings on the windward side of the building model and divide it into grids.

[0086] S2: Define a porous medium region at the opening of the building model and make the porous medium region cover the opening of the building model.

[0087] In this embodiment, a porous medium block model with the same opening size is placed in the porous medium region.

[0088] S3: Add a drag source term to the porous medium region so that the drag coefficient is greater than or equal to a set threshold under initial conditions; during the numerical simulation, make the drag coefficient suddenly equal to zero or decrease to zero according to a set curve to achieve the simulation effect of a sudden opening. This includes the following steps:

[0089] S31: Adding a resistance source term: The porous medium block model is constructed based on Darcy-Forchheimer theory. The momentum equation of the porous medium block model after adding a resistance source term is:

[0090]

[0091]

[0092] Wherein: S m The momentum source term is represented in the direction m (m = x, y, z); μ represents the hydrodynamic viscosity; ρ represents the fluid density; σ represents the stress tensor; U m |U| represents the fluid velocity in the direction m (m = x, y, z); |U| represents the velocity matrix in the three directions; D and F are the viscous drag coefficient and inertial drag coefficient in the Darcy-Forchheimer model, respectively.

[0093] S32: Sudden Opening: Under initial conditions, the viscous drag coefficient D and the inertial drag coefficient F are made greater than or equal to a set threshold. During the numerical simulation, the viscous drag coefficient D and the inertial drag coefficient F are suddenly made equal to zero or decrease to zero according to a set curve to achieve the simulation effect of a sudden opening. Specifically, the set threshold for the viscous drag coefficient D and the inertial drag coefficient F under initial conditions can be 10^15. At a certain moment in the simulation (e.g., when t equals 2s), the viscous drag coefficient D and the inertial drag coefficient F are changed to 0 to achieve the effect of a sudden opening.

[0094] Therefore, by creating porous media blocks with the same size as the openings at the openings of the building model, and by adjusting the resistance value of the porous media blocks, the closure and opening of the building can be simulated. Figure 2 (a)-(b) represent axial slice diagrams of meshes with and without defined porous media.Figure 2 (b) The red area is a porous medium area, the size of which is the same as the opening size, thus ensuring that the building model is completely sealed.

[0095] In this embodiment, a model with an opening ratio of 3.1% is selected for the study, that is, an opening size of 20×20 (mm). 2 Assuming that the porous medium resistance value is changed to 0 at time 2 seconds after the calculation begins, the simulation will show a sudden opening.

[0096] Figure 3 Images (a)-(d) show two-dimensional slices along the y=0 axis of the simulation results using the porous media method, illustrating the instantaneous combined velocity and pressure fields inside and outside the model 0.5 s before and after the opening. It can be observed that at 2 s after the opening, the porous media effectively impedes airflow into the building; the airflow velocity inside the model before the opening is 0. As the drag coefficient becomes 0, airflow quickly enters near the opening inside the building, and the internal pressure becomes positive. However, the porous media cannot prevent pressure transmission; that is, the internal pressure of the model is abnormally negative just before the opening, whereas in reality, the pressure inside a closed model building is usually zero.

[0097] Therefore, to investigate the specific impact of whether the initial pressure inside the building at the moment of opening is zero on the internal pressure of the structure, the saved flow field information file at the moment of opening (i.e., time 2 seconds) was used. The pressure in the grid region inside the building model was reset to 0 using the setFields tool built into OpenFoam. Flow field calculations were then performed based on these two initial conditions. Furthermore, a monitoring point was set up inside the building near the opening to monitor the airflow at the orifice and the pressure fluctuations inside the building.

[0098] Figure 4 The time history diagrams of the internal pressure coefficient and longitudinal wind speed (U) at the orifice of the building are shown when the initial internal pressure values ​​are -260 Pa and 0 Pa, respectively. Figure 4 (a)-(b) show that different initial pressure values ​​at the moment of opening affect the wind speed and pressure within a very short time (within 0.003s) at the instant of opening. The peak internal pressure coefficient under a negative initial pressure can reach approximately 9, which is highly unusual. However, after resetting the initial internal pressure to 0, the peak internal pressure coefficient at the instant of opening is around 1.5. After a period of time, the simulated internal pressure results under both different initial conditions show the same trend. Similarly, in... Figure 4As shown in (b), for the condition with an initial negative pressure, negative flow appears near the orifice at the instant of opening, meaning the airflow flows from the inside of the building to the outside. This is likely due to the initial negative pressure inside the building, which is inconsistent with the actual phenomenon of airflow flowing into the building when the windward side is opened. After a period of time, the change trend of the simulated orifice velocity under the two different initial conditions is the same. When the initial value is reset to 0 internal pressure, it returns to normal.

[0099] Example 2: Numerical Simulation Method for Internal and External Pressure Changes under Sudden Building Openings

[0100] The numerical simulation method for changes in internal and external pressure under sudden openings in a building, as described in this embodiment, includes the following steps.

[0101] Step 1: Construct a building model with a sudden opening.

[0102] 11) Use modeling software to create a square building model, open an opening on the windward side of the building model and divide it into grids.

[0103] 12) Define a porous medium region at the opening of the building model and make the porous medium region cover the opening of the building model; in this embodiment, a porous medium block model with the same size as the opening is placed at the porous medium region.

[0104] 13) Add a drag source term to the porous medium region so that the drag coefficient is greater than or equal to a set threshold under initial conditions; during the numerical simulation, make the drag coefficient suddenly equal to zero or decrease to zero according to a set curve to achieve the simulation effect of a sudden opening. This includes the following steps:

[0105] 131) Adding a resistance source term: The porous medium block model is constructed based on Darcy-Forchheimer theory. The momentum equation of the porous medium block model after adding a resistance source term is:

[0106]

[0107] Wherein: S m The momentum source term is represented in the direction m (m = x, y, z); μ represents the hydrodynamic viscosity; ρ represents the fluid density; σ represents the stress tensor; U m |U| represents the fluid velocity in the direction m (m = x, y, z); |U| represents the velocity matrix in the three directions; D and F are the viscous drag coefficient and inertial drag coefficient in the Darcy-Forchheimer model, respectively.

[0108] 132) Sudden Opening: The viscous drag coefficient D and the inertial drag coefficient F are initially set to be greater than or equal to a set threshold. During the numerical simulation, the viscous drag coefficient D and the inertial drag coefficient F are suddenly made equal to zero or decrease to zero according to a set curve to achieve the simulation effect of sudden opening. Specifically, the set threshold for the viscous drag coefficient D and the inertial drag coefficient F under initial conditions can be 10^15. At a certain moment in the simulation (e.g., when t equals 2s), the viscous drag coefficient D and the inertial drag coefficient F are changed to 0 to achieve the effect of sudden opening.

[0109] Specifically, the TPU database is a comprehensive database jointly established by Tokyo Polytechnic University. This embodiment uses the wind tunnel pressure measurement experiment of a low-rise building under atmospheric flow from the TPU database. The wind field topography type corresponds to Class III terrain in the Japanese Wind Load Code 2004, with a mean wind speed profile index of 0.2, a reference height of 10m at full size, and a corresponding turbulence intensity of 25%. The model selected is a rigid flat roof model with an aspect ratio of 1:1 and a width-to-height ratio of 2:1, with dimensions of 160mm × 160mm × 80mm (L × B × H) and a geometric scale ratio of 1:100. For ease of description, this embodiment refers to this model as the TPU model.

[0110] Since subsequent monitoring of the pressure distribution on the inner and outer pressure surfaces of the opening structure is necessary, this embodiment references existing experimental models for studying wind-induced opening internal pressure. The TPU model has been modified accordingly, featuring a double-sided sandwich structure to ensure proper arrangement of the pressure measuring pipes. The sandwich layer is treated as a wall surface, with a thickness of 10mm. Therefore, the actual internal dimensions of the model are 140×140×60 (mm). 3 The building model used in this study has a square opening. For ease of representation, the side length of the opening is denoted as l0. Specifically, in this embodiment, the side length of the opening, l0, is 0.02m. Figure 5 The diagrams show geometric and cross-sectional views of models of both enclosed and open buildings.

[0111] Step 2: Design the computational domain

[0112] like Figure 6As shown, this embodiment constructs a computational domain with dimensions of 30H×10H×6H (H=80mm). The building model is placed within the computational domain, with the origin of the domain located at the center of the bottom of the building model. The distances from the center of the bottom of the building model to the inlet and outlet of the computational domain are 10H and 20H, respectively, where H is the height of the building model. This ensures sufficient development of the inlet flow. The distance from the inlet to the center of the model is 10H. To meet the requirements of building wind tunnel testing, the model's blockage rate is 3.3%, satisfying the requirement that the blockage effect is less than 5%. The same computational domain size is used whether the building model's opening is closed or open.

[0113] Step 3: Grid Generation

[0114] The building model and the computational domain are meshed separately. To ensure that the boundary flow near the wall of the building model is solved accurately, a boundary layer region is set at the wall of the building model.

[0115] In this embodiment, three mesh schemes were used for TPU model verification: coarse mesh, medium mesh, and fine mesh for mesh independence analysis, denoted as mesh 1, mesh 2, and mesh 3, as shown in Table 1. The computational domain was preprocessed using the OpenFoam preprocessing tools blockMesh and snappyHexMesh, including mesh generation, densification, and boundary layer addition. The three mesh schemes were generated by changing the number of meshes in the x, y, and z directions within the blockMesh tool. The entire computational domain was drawn using a structured mesh, with unstructured meshes used between different density transition zones. Four density levels of mesh regions were established, achieving a gradual transition from large to small mesh sizes. The diagonal coordinates of the first-level density region were (-10H, -2.5H, 0) and (5H, 2.5H, 2H); the diagonal coordinates of the second-level density region were (-2H, -1.5H, 0) and (2H, 1.5H, 1.5H); and the diagonal coordinates of the third-level density region were (-1.125H, -1.125H, 0) and (1.125H, 1.125H, 1.125H). A detailed density diagram can be found below. Figure 7 .

[0116] Near the building target, the mesh size is very important for the simulation accuracy. To ensure that the boundary flow near the wall is solved accurately, a boundary layer region needs to be set at the model wall. Figure 8 (a)-(b) show the slices of the refined mesh along the y=0 direction and the schematic diagram of the boundary layer mesh when the model opening is closed. A dimensionless distance y+ is introduced to quantify the flow state of the fluid near the wall. The mesh refinement scheme of mesh 2 ensures that the average y+ value of the first layer of mesh near the wall basically meets the simulation requirements, as shown in Table 1.Figure 8 (c)-(d) also show schematic diagrams of slices of the refined mesh along the y=0 direction and local meshes near the opening under the building model. The watershed mesh inside the entire model is approximately 600,000, and the total number of meshes is approximately 3.4 million. When the model opening is open, compared with the mesh computation domain when the opening is closed, except for the mesh near the building opening, the mesh size is the same in the other refined areas.

[0117] Table 1 Mesh Scheme for Closed Model

[0118]

[0119] Step 4: Set boundary conditions

[0120] Inlet boundary conditions: Atmospheric inflow boundary conditions are used at the inlet, and a consistent discrete inflow turbulence generation method is employed. Outlet boundary conditions: The outlet boundary conditions are set as a pressure outlet. Slip boundary conditions are set at the top and both sides of the computational domain, while no-slip boundary conditions are set at the bottom of the computational domain and the building wall.

[0121] Specifically, the initial physical field boundary conditions for the numerical example in this embodiment are as follows: the inlet adopts atmospheric inflow boundary conditions, and the consistent discrete inflow turbulence generation method, i.e., the CDRFG method, is used. The wind profile data from the TPU wind tunnel test used in this embodiment are shown in Table 2.

[0122] Table 2 Parameter settings for the inlet turbulent wind field

[0123]

[0124] Note: According to the European standard (ESDU 85020), some parameters in the table have Z0 = 0.7m.

[0125] For a detailed schematic diagram of the flow field boundary in the computational domain, please refer to [link / reference]. Figure 7 The outlet boundary condition is set to pressure outlet; the top and two sides of the computational domain are set to slip boundary conditions, which means that the tangential velocity of the fluid passing through the boundary surface is not zero and the velocity direction is parallel to the boundary surface; the bottom of the computational domain and the building wall boundary conditions are set to no-slip boundary conditions, which means that the fluid does not pass through the boundary surface and all its normal velocity components with the boundary surface are zero.

[0126] Turbulent inlet typically refers to atmospheric boundary layer (ABL) inflow. In numerical simulations, the goal is to generate a numerical turbulent wind field at the inlet interface with statistical characteristics similar to real-world natural wind fields. This requires including mean wind speed profiles, turbulence intensity, fluctuating wind speed power spectra, and turbulence integral scales, among other things.

[0127] In this embodiment, the mathematical expression for the fluctuating wind speed synthesized using the Consistent Discrete Incoming Turbulence Generation Method (CDRFG) is as follows:

[0128]

[0129] Where: u i i = 1, 2, 3 represent the fluctuating wind speeds in the x, y, and z directions, respectively; j = 1, 2, 3 represent the x, y, and z directions, respectively; M represents the number of frequency intervals in the spectrum; N represents the number of frequencies in each interval. and All are coefficients; It represents a random vector distributed on a unit sphere in space, ensuring that the velocity field satisfies the divergence-free condition; Represents a dimensionless coordinate vector, and f is the correlation coefficient between two points in space; n,m Let f be the frequency, and let f be the mean. m The standard deviation Δf is a uniform distribution; t is time; and:

[0130]

[0131] Where: sign is the sign function; S represents a random number that follows a 0-1 normal distribution in all three directions; ui (f m ) is the spectrum f m The corresponding power spectral density value in the i-direction.

[0132] This embodiment uses an artificial synthesis method to generate the inlet atmospheric flow. The specified parameters for the generated turbulent wind field at the inlet are shown in Table 2. To ensure the accuracy of the wind load on the target building surface, a wind profile consistent with the TPU experiment needs to be monitored at the building location (i.e., the center of the computational domain origin) under an open wind field. However, since the pulsating wind speed synthesized in CDRFG only satisfies some characteristic conditions of the atmospheric flow on some parameters, when the wind field enters the computational domain, the turbulent wind field monitored at the building location will exhibit a certain degree of nonlinear dissipation in the height direction. Secondly, this method cannot completely guarantee that the turbulent inlet strictly meets the dispersion condition, which can also lead to abnormal virtual pressure fluctuations in the numerical calculation. Therefore, this embodiment introduces an improved "Iterating-Reshaping" method (IRCDRFG) to continuously correct the shape of the wind speed profile at the inlet. This ensures that the turbulent wind field can fit the target wind profile within the target area without affecting spatial correlation, and maintains the conservation of flow through the inlet interface at all times. Specifically, the principle of the iterative reshaping method is as follows:

[0133]

[0134] In the formula: U mean,t (z) and I i,t (z) represents the target mean velocity profile and the target turbulence intensity profile; and It is the average wind speed profile and turbulence intensity profile obtained from a simulated time history at the center of the building area during the nth correction period; and These are the mean wind speed profile and turbulence intensity profile of the turbulent field generated at the inlet interface during the nth correction; γ n Δz is a correction factor used to ensure that the flow rate through the inlet interface remains constant after the (n+1)th correction of the average wind speed profile at the inlet interface; Δz is the grid scale corresponding to the z-direction.

[0135] Figure 9 (a)-(b) show the wind speed profile at the building target and the dimensionless power spectrum at the reference height after four iterations of optimization based on the "Iterative Reshaping (IRCDRFG)" method under the air-wind field. It can be seen that the average wind speed and turbulence intensity of the numerical monitoring are consistent with the target values ​​of the TPU wind tunnel experiment, and the wind speed power spectrum at the reference height maintains good consistency with the Karman spectrum, indicating that the turbulent wind field generated based on IRCDRRFG is accurate and reliable.

[0136] Step 5: Numerical Simulation

[0137] The simulation is based on the large eddy turbulence model. Under the initial conditions, the drag coefficient of the porous medium region of the building model is made to be greater than or equal to a set threshold to close the opening. After the wind speed and pressure of the whole watershed stabilize, the drag coefficient of the porous medium region of the building model is set to 0 to open the opening. This realizes the simulation of the sudden opening of the building.

[0138] In this embodiment, the numerical solution settings are as follows: the open-source software OpenFoam is used, the LES turbulence model is adopted, the subgrid-scale filtering function is selected as the WALE subgrid model according to the literature, and the relevant turbulence model coefficients are: C e =1.048, C k =0.094, C w =0.544, kinematic viscosity is 1.455×10 -5 m 2 / s. In the discretization scheme settings, a second-order backward difference scheme was selected for time discretization. The diffusion and convection terms in the governing equations were handled using a second-order central difference scheme and a linear upwind stable transport scheme (LUST), respectively. Higher-order discretization schemes were chosen to ensure the accuracy of the numerical simulation. In the solver settings, a geometric algebraic multigrid (GAMG) solver and a preprocessed bijugate gradient (PBiCG) solver were used for pressure and velocity solutions, respectively. The convergence residual for iterative calculations was set to 1×10⁻⁶. -6 The Pimple algorithm was selected for transient solutions to velocity-pressure coupling problems.

[0139] In selecting the simulation time step, the specific impact of the time step on simulation accuracy is evaluated. This is especially relevant when the dimensionless time step t... * When tU0 / D (t is the simulation time step) does not exceed 0.025, the simulation results show consistency within this range. U0 is the average wind speed at the reference height, and D is the model characteristic length. In this embodiment, the dimensionless time step is 0.00875, which meets the requirements. The ratio of the entire calculation basin length to the wind speed at the reference height is called the full basin time. To obtain a stable convergent solution, preliminary calculations show that the wind speed and wind pressure are basically stable after 5 full basin time monitoring periods. Therefore, this embodiment selects the calculation results from the 6th to the 47th full basin time periods for statistical analysis, with the corresponding dimensionless time length tI. H The value of / H is approximately 1400, and the maximum Courland number (CFL) in the calculation is approximately 1.2, which ensures convergence while also meeting the 10-minute statistical time requirement in the wind tunnel test specifications.

[0140] In this embodiment, the Large Eddy Turbulence (LES) model from numerical simulation is used for numerical calculations. The LES model, first proposed by Smagorinsky, lies between the Reynolds-averaged flow model (RANS) and direct simulation (DNS). It divides the eddy scale in the flow field into large-scale and small-scale eddies using a spatial filtering function. Specifically, the method for simulating and solving the Large Eddy Turbulence model is as follows:

[0141] Filter all vortex scales Φ(x,t) in the flow field:

[0142]

[0143] in: For large-scale vortices obtained through filtering; Φ′(x,t) is the subgrid scale; G(x,x′) is the spatial filtering function; and:

[0144]

[0145] Where: x is the spatial coordinate before filtering; x′ is the spatial coordinate after filtering; Δ is the filtering scale;

[0146] The filtered incompressible fluid control equations are as follows:

[0147]

[0148] in: (i, j = 1, 2, 3) are the three filtered velocity components in the Cartesian coordinate system; ρ is the filtered pressure; ν is the air density; ν is the kinematic viscosity; τ is the pressure after filtering. ij Represents subgrid-scale stress;

[0149] Small-scale vortices are modeled using a subgrid-based eddy viscosity model, where the expression for the standard SGS model is:

[0150]

[0151]

[0152] Where: μ t It is the SGS eddy viscosity coefficient, δ ij It is the Kronecker function. It is the filter thickness strain rate tensor;

[0153] Eddy viscosity coefficient μ t Calculated using the Smagorinsky-Lilly model:

[0154]

[0155] Where: k is the von Karman constant, usually taken as 0.42; C s Δ is the Smagorinsky constant, typically 0.1; Δ is the filtering scale.

[0156] Step Six: Numerical Analysis

[0157] Numerical analysis was performed on the pressure and wind speed data inside and outside the building collected under the condition of a sudden opening in the building model.

[0158] 6.1 Working conditions and measurement point layout of the building model

[0159] When the model opening is closed, this embodiment simulates the wind pressure measurement point layout of the TPU low-rise building experiment. The roof pressure measurement points are arranged in the same way as the TPU test pressure measurement points, with a total of 64 pressure measurement points. Secondly, to compare the test results, a total of 40 pressure measurement points are also arranged along the central axis of the building, with increased density along the line. See [link to relevant documentation]. Figure 10After the model opening is opened, this embodiment also arranges 8 external pressure measuring points around the opening, denoted as W1-W8. Inside the building model, on the windward, leeward, crosswind, and roof surfaces, 5 internal pressure measuring points are set at the center of the wall, denoted as N1-N5. (See...) Figure 11 Furthermore, to compare the difference in external pressure distribution between the open and closed models, two working conditions were set for the closed TPU model at 0° and 45° angles, with the mesh refinement method being the same as... Figure 7 Consistent.

[0160] This embodiment studies the building model under different wind angles, totaling eight operating conditions. Specific parameters for each condition are detailed in Table 3. The numerical model for each condition is set up in the same way as the closed structure example, and the inlet turbulence conditions are consistent.

[0161] Table 3 Operating Condition Settings

[0162]

[0163] 6.2 Spatial distribution of pressure within the model

[0164] When building openings are closed, for pressure measurement points on the roof and along the building's central axis, each pressure measurement point is ensured to be located within the first-floor height of the near-wall grid. The wind pressure data is processed using the following dimensionless wind pressure coefficient formula:

[0165]

[0166] In the formula: p(t) is the pressure monitoring time history at the pressure measuring point; p0 is the static pressure at the reference point, the reference pressure measured at the inlet center; ρ is the air density, 1.225 kg / m³. 3 ;U ref To reference the average wind speed at the eaves height, the model height H is 0.08m, and the average wind speed in the TPU numerical simulation is taken as 13.5m / s. Average wind pressure coefficient... Pulsating wind pressure coefficient C' p and extreme wind pressure coefficient These correspond to the average value, standard deviation, and extreme value of the wind pressure coefficient, respectively.

[0167] When a building has openings, considering background leakage and the opening itself, the relationship between the average internal wind pressure coefficient and the average external wind pressure coefficient of the openings under multiple openings is derived based on Bernoulli's theorem and the law of conservation of mass:

[0168]

[0169] In the formula: A W Indicates the size of the opening area on the windward side; A L Indicates the size of the area where the background is leaked; This indicates the average wind pressure coefficient inside the building; This indicates the average wind pressure coefficient outside the building opening; This represents the average external wind pressure coefficient at the orifice where there is a background leak. Therefore, for a single opening with no leak, we can obtain:

[0170]

[0171] The external wind pressure coefficient of the opening structure is obtained by taking the average time history of the measuring points (W1-W8) around the orifice that are highly correlated with the internal pressure. The average external wind pressure coefficient is then used to derive the theoretical value of the internal pressure inside the single opening structure. This value is then compared with the simulated values ​​at the internal pressure measuring points (N1-N5) to verify the accuracy of the internal pressure of the single opening structure. Figure 12 The numerical simulation results of the internal pressure under a single open structure facing the wind are presented, and the theoretical value of the internal pressure is compared with that of the formula. As can be seen from the figure, the result estimated by the theoretical equation of the average internal pressure coefficient is in good agreement with the numerical simulation result.

[0172] Depend on Figure 12 (a) It can be seen that the average internal pressure of the open structure varies with the opening ratio in the range of 0.69-0.74, and the average internal pressure does not change much when the opening increases to a certain extent. At 0° angle, the numerical results for different opening ratios are in better agreement with the theoretical results than those for larger opening ratios, with a maximum error of about 9%. This may be because the external pressure measurement points and internal pressure measurement points around the orifice are more correlated when the opening is small, so the estimate of the average internal pressure coefficient is relatively accurate.

[0173] Depend on Figure 12 (b) It can be seen that, for the same opening ratio but different wind angles, the numerical results of the average internal pressure coefficient are in good agreement with the theoretical results, with a maximum error of about 5%. Furthermore, the average internal pressure changes with the wind angle, first decreasing and then increasing, with a range of -0.61 to 0.73. In addition, it can be found that for a single opening at a wind angle of 0° on the windward side, the average internal pressure coefficient is the highest under this condition. Therefore, a wind angle of 0° is the most unfavorable location for a single-opening building.

[0174] Numerical results show that for examples with different opening sizes and wind angles, the average internal pressure simulated by LES in this embodiment can fit the theoretical value well, and the maximum error does not exceed 10%. Therefore, it can be considered that the numerical scheme for the TPU opening model is applicable.

[0175] 6.3 Distribution of External Pressure Around the Model

[0176] Furthermore, to investigate the distribution law of external pressure on the surface of the building model, the surface wind pressure of the model in the working case example was processed using the dimensionless wind pressure coefficient formula to obtain the surface wind pressure coefficient. Subsequently, averaging and variance processing were performed to obtain the average wind pressure coefficient and the fluctuating wind pressure coefficient of the model surface. The aperture ratio (the ratio of the opening area to the area of ​​the wall where the opening is located) is denoted as ρ.

[0177] Figure 13 (a)-(d) are cloud maps showing the average wind pressure coefficient distribution around the outer perimeter of a closed model with ρ = 0.3%, ρ = 3.1%, and ρ = 12.5% ​​at a wind direction angle of 0 degrees. Figure 12 From (a) to (d), it can be observed that, compared to the closed model, the maximum positive pressure area on the windward side of models with different opening ratios gradually decreases as the opening ratio increases. For the model with ρ = 12.5%, the area with the maximum average wind pressure coefficient on the windward side is even smaller. This may be because, with large openings in the building, airflow can more easily enter the building interior, forming a backflow at the opening, further reducing the airflow velocity acting on the wall. It is worth noting that, apart from the front area where the building model is located, the negative pressure areas on the roof and sides do not change significantly.

[0178] Figure 14 (a)-(b) present the distribution of fluctuating wind pressure coefficients on the external surface of the closed model, with ρ = 0.3%, ρ = 3.1%, and ρ = 12.5% ​​at a wind direction angle of 0°. The observations show that the distribution area of ​​the fluctuating wind pressure coefficient remains consistent between the closed and small opening ratio models, and the fluctuating wind pressure area around the opening does not increase significantly. In contrast, the high fluctuating wind pressure area near the opening of the large opening ratio model increases significantly, and the high fluctuating areas on the roof and sides are significantly larger than those of the closed model. This further illustrates the complexity of the flow around the opening under large opening conditions, increasing the area of ​​high fluctuating wind pressure coefficients on the surface where the building opening is located.

[0179] from Figures 13-14 It can be observed that the impact of a single opening ratio on wind pressure on the windward side is mainly concentrated on the windward side where the opening is located, with a smaller impact on the sides and roof. To quantify the impact of the opening ratio on the external pressure distribution of low-rise buildings, the wind pressure coefficient distribution along the central axis of the model is taken, and the opening parts of the model are replaced by dashed lines. Figure 15 The paper shows the changes in the average wind pressure coefficient and fluctuating wind pressure coefficient along the central axis of the closed model at 0° angle and the models with different opening ratios. Figure 15(a) It can be observed that at the windward side, the positive pressure range for the closed model is 0.65 to 0.78. However, as the opening ratio increases, the positive pressure at the windward side decreases, but the fluctuation is more pronounced. Specifically, the positive pressure fluctuation range for the open model with an opening ratio ρ = 12.5% ​​is -0.54 to 0.72, a reduction of approximately 15% compared to the closed model. In the flow separation area at the corner of the roof, the negative pressure decreases with increasing opening ratio, but only the open model with an opening ratio ρ = 12.5% ​​shows an increase of approximately 8% compared to the closed model. For the other opening ratio models, there is no significant change in wind pressure distribution on the roof and in other surrounding areas. Figure 15 (b) It can be observed that, compared with the closed model, as the opening ratio increases, the fluctuating wind pressure coefficient near the windward face is slightly higher than that of the closed model, but the fluctuation range is not too large. The fluctuating wind pressure coefficient on the windward face in all operating conditions is within the range of 0.35-0.45. It is worth noting that the fluctuating wind pressure in the flow separation area at the front edge of the roof of the model with an opening ratio of 3.1% is significantly smaller than that of the closed model. Apart from this, the fluctuating wind pressure distribution on the roof and leeward face of the other open models does not change much.

[0180] Figure 16 (a)-(b) show the contour maps of the average wind pressure on the exterior of the low-rise building under a 45° wind angle for the closed and open models. It can be seen that at a 45° angle, the average wind pressure distribution on the roof of both the closed and open models does not change significantly, and the roof wind pressure exhibits a conical distribution, showing clear flow separation. However, in the corner region of the windward side of the model, the average positive pressure area of ​​the open model is larger than that of the closed model, while the average wind pressure area on the surface where the opening is located is smaller than that of the closed model. This may be because the airflow near the opening reduces the incoming wind speed acting on the surface near the opening, resulting in a decrease in pressure.

[0181] Figure 17 (a)-(b) show the contour maps of the average wind pressure outside the low-rise building under a 45° wind angle, representing the closed-off model with an open section. Figure 16 It can be seen that the distribution of the roof pulsation area in the open model is basically consistent with that in the closed model at the corresponding 45° angle. However, the open model has a significant reduction in the high pulsation area on the two windward sides, and the surface pulsation wind pressure on the open side is smaller than that in the closed model.

[0182] Note: In this embodiment, "open model" refers to the model when the opening of the building model is open, and "closed model" refers to the model when the opening of the building model is closed.

[0183] The above-described embodiments are merely preferred embodiments provided to fully illustrate the present invention, and the scope of protection of the present invention is not limited thereto. Equivalent substitutions or modifications made by those skilled in the art based on the present invention are all within the scope of protection of the present invention. The scope of protection of the present invention is defined by the claims.

Claims

1. A numerical simulation method for changes in internal and external pressure under sudden openings in a building, characterized in that: Includes the following steps: Step 1: Construct a building model with a sudden opening. 11) Use modeling software to create a square building model, open an opening on the windward side of the building model and divide it into a grid; 12) Define a porous medium region at the opening of the building model and make the porous medium region cover the opening of the building model; 13) Add resistance source terms to the porous media region; Step 2: Design the computational domain The size of the building is 30. H ×10 H ×6 H The computational domain is defined, and the building model is placed within it, with the origin of the computational domain located at the center of the bottom of the building model. The distances from the center of the bottom of the building model to the entrance and exit of the computational domain are 10. H and 20 H , H The height of the building model; Step 3: Grid Generation The building model and the computational domain are meshed separately. To ensure that the boundary flow near the wall of the building model is solved accurately, a boundary layer region is set at the wall of the building model. Step 4: Set boundary conditions Inlet boundary conditions: Atmospheric inflow boundary conditions are used at the inlet, and the uniform discrete inflow turbulence generation method is employed; Export boundary conditions: The export boundary conditions are set to pressure outlet; The top and two sides of the computational domain are set as sliding boundary conditions, while the bottom of the computational domain and the building wall boundary conditions are set as non-slip boundary conditions. Step 5: Numerical Simulation The simulation was performed based on the large eddy turbulence model. Under the initial conditions, the drag coefficient of the porous medium region of the building model was made to be greater than or equal to a set threshold in order to close the opening. Once the wind speed and pressure across the entire basin have stabilized, the drag coefficient of the porous medium region of the building model will be set to 0 to achieve the purpose of opening the opening. Simulate a sudden opening in a building; Step Six: Numerical Analysis Numerical analysis was performed on the pressure and wind speed data inside and outside the building collected under a sudden opening in the building model. In step 12), a porous medium block model with the same opening size is placed in the porous medium region; Step 13) also includes the following steps: 131) Adding a resistance source term: The porous medium block model is constructed based on Darcy-Forchheimer theory. The momentum equation of the porous medium block model after adding a resistance source term is: in: This represents the momentum source term in the direction of m (m = x, y, z); Fluid dynamic viscosity; Indicates fluid density; Represents the stress tensor; This represents the fluid velocity in the direction m (m = x, y, z); Represents the velocity matrix in three directions; and These are the viscous drag coefficient and inertial drag coefficient in the Darcy-Forchheimer model, respectively. 132) Sudden opening: reduces the viscous resistance coefficient With inertial drag coefficient Under initial conditions, the coefficient of viscous resistance should be greater than or equal to a set threshold; during numerical simulation, the coefficient of viscous resistance should be... With inertial drag coefficient The equal parts suddenly become zero or decrease to zero according to a set curve to achieve the simulation effect of a sudden opening.

2. The numerical simulation method for internal and external pressure changes under sudden openings in a building according to claim 1, characterized in that: In step four, the mathematical expression for the fluctuating wind speed synthesized using the consistent discrete incoming flow turbulence generation method is as follows: in: They represent x , y , z Pulsating wind speeds in three directions; They represent x , y , z Three directions; M Indicates the number of intervals in the spectrum; N This indicates the number of frequencies within each interval. and All are coefficients; It represents a random vector distributed on a unit sphere in space, ensuring that the velocity field satisfies the divergence-free condition; Represents a dimensionless coordinate vector, and , The correlation coefficient between two points in space; Let be the frequency, and follow the mean. Standard deviation For a uniform distribution of ; t is time; and: in: It is a symbolic function; This represents a random number that follows a normal distribution of 0-1 in all three directions. For the spectrum corresponding i Power spectral density value in the direction.

3. The numerical simulation method for internal and external pressure changes under sudden openings in a building according to claim 2, characterized in that: In the uniform discrete incoming flow turbulence generation method, an iterative reshaping method is introduced to continuously correct the shape of the wind speed profile at the inlet. This ensures that the turbulent wind field can fit the target wind profile within the target region without affecting spatial correlation, while maintaining the conservation of flow rate through the inlet interface. The principle of the iterative reshaping method is as follows: In the formula: and These are the target mean velocity profile and the target turbulence intensity profile; and It is the first n During the secondary correction period, the average wind speed profile and turbulence intensity profile were obtained by monitoring at the center of the building area after a simulated time history. and It is in the n The average wind speed profile and turbulence intensity profile of the turbulent field generated at the inlet interface during the second correction; This is a correction factor used to ensure that the average wind speed profile at the inlet interface is in the ( ) n After the +1) correction, the traffic through the entry interface remains constant. yes z The grid scale corresponding to the direction.

4. The numerical simulation method for internal and external pressure changes under sudden openings in a building according to claim 1, characterized in that: In step five, the method for simulating and solving the large eddy turbulence model is as follows: All vortex scales in the flow field Perform filtering: in: This refers to large-scale vortices obtained through filtering. Subgrid scale; Let be a spatial filtering function; and: in: These are the spatial coordinates before filtering; Filtered spatial coordinates; The filtering scale; The filtered incompressible fluid control equations are as follows: in: ( i , j =1,2,3) are the three filtered velocity components in the Cartesian coordinate system; It is the pressure after filtering; ρ It is air density; ν It is kinematic viscosity; Represents subgrid-scale stress; Small-scale vortices are modeled using a subgrid-based eddy viscosity model, where the expression for the standard SGS model is: in: It is the SGS eddy viscosity coefficient. It is the Kronecker function. It is the filter thickness strain rate tensor; Eddy viscosity coefficient Calculated using the Smagorinsky-Lilly model: in: k It is von Karman's constant; This is Smagorinsky's constant; This is the filtering scale.

Images